10.1261/rna.033530.112Access the most recent version at doi: 2012 18: 1553-1562 originally published online June 15, 2012RNA

 Kristin Benjamin Talsky and Kathleen Collins uridylyltransferase protein partnerproduction requires a previously uncharacterized

small RNATetrahymenaStrand-asymmetric endogenous   

References http://rnajournal.cshlp.org/content/18/8/1553.full.html#ref-list-1

This article cites 20 articles, 10 of which can be accessed free at:

serviceEmail alerting

click heretop right corner of the article orReceive free email alerts when new articles cite this article - sign up in the box at the

http://rnajournal.cshlp.org/subscriptions go to: RNATo subscribe to

Copyright © 2012 RNA Society

Cold Spring Harbor Laboratory Press on July 16, 2012 - Published by rnajournal.cshlp.orgDownloaded from

Strand-asymmetric endogenous Tetrahymena small RNA

production requires a previously uncharacterizeduridylyltransferase protein partner

KRISTIN BENJAMIN TALSKY and KATHLEEN COLLINS1

Department of Molecular and Cell Biology, University of California, Berkeley, California 94720-3200, USA

ABSTRACT

Many eukaryotes initiate pathways of Argonaute-bound small RNA (sRNA) production with a step that specifically targets sets ofaberrant and/or otherwise deleterious transcripts for recognition by an RNA-dependent RNA polymerase complex (RDRC). Thebiogenesis of 23- to 24-nt sRNAs in growing Tetrahymena occurs by physical and functional coupling of the growth-expressedDicer, Dcr2, with one of three RDRCs each containing the single genome-encoded RNA-dependent RNA polymerase, Rdr1.Tetrahymena RDRCs contain an active uridylyltransferase, either Rdn1 or Rdn2, and Rdn1 RDRCs also contain the Rdf1 andRdf2 proteins. Although Rdn2 is nonessential and RDRC-specific, Rdn1 is genetically essential and interacts with a non-RDRCprotein of 124 kDa. Here we characterize this 124-kDa protein, designated RNA silencing protein 1 (Rsp1), using endogenouslocus tagging, affinity purification, and functional assays, as well as gene-knockout studies. We find that Rsp1 associates withRdn1-Rdf1 or Rdn1-Rdf2 subcomplexes as an alternative to Rdr1, creating Rsp1 complexes (RSPCs) that are physically separatefrom RDRCs. The uridylyltransferase activity of Rdn1 is greatly reduced in RSPCs compared with RDRCs, suggesting enzymeregulation by the alternative partners. Surprisingly, despite the loss of all known RDRC-generated classes of endogenous sRNAs,RSP1 gene knockout was tolerated in growing cells. A minority class of Dcr2-dependent sRNAs persists in cells lacking Rsp1with increased size heterogeneity. These findings bring new insights about the essential and nonessential functions of RNAsilencing in Tetrahymena, about mechanisms of endogenous small interfering RNA production, and about the roles of cellularuridylyltransferases.

Keywords: RNA silencing; RNA-dependent RNA polymerase; poly(U) polymerase; Piwi RNP

INTRODUCTION

Among the cellular RNA recognition mechanisms thatcompete to determine the fate of mRNAs and mRNA-liketranscripts, the specificity for initiating endogenous RNAsilencing is among the least understood. Production of someendogenous small RNAs (sRNAs) from annealed transcriptsis thought to be initiated by double-stranded RNA (dsRNA)binding to Dicer complexes (Carthew and Sontheimer 2009;Ketting 2011). However, in many eukaryotes, the abundantendogenous sRNAs are dependent on active rather than passivegeneration of dsRNA mediated by an RNA-dependent RNApolymerase (Rdr). Some Rdr-dependent dsRNA synthesisoccurs secondary to original transcript targeting for RNAsilencing, for example, the production of some types of small

interfering RNA (siRNA) in Arabidopsis thaliana and Dicer-independent amplification of siRNAs in Caenorhabditis elegans(Vazquez et al. 2004; Pak and Fire 2007; Sijen et al. 2007). Rdr-dependent dsRNA synthesis can also be specified by tetheringthe enzyme to chromatin, as established in Schizosaccharomycespombe in mechanistic detail (Grewal 2010; Moazed 2011). Inaddition to these functions in perpetuating siRNA production,an Rdr can act upstream of any other known pathway step toinitiate the silencing of mRNA-like transcripts (Voinnet 2008;Ketting 2011; Thivierge et al. 2012). The principles thatunderlie the cellular specificity of transcript recognition byan Rdr are unknown, but they must provide for both targetgenerality (to be evolutionarily useful) and also target speci-ficity (to avoid selecting a functional RNA for self-perpetuatingsilencing). It seems likely that the varied subunits of cellularRdr complexes (RDRCs) contribute to the specialization ofRdr function in vivo, restricting the general activity of Rdrenzymes that is observed on single-stranded RNA templatesin vitro (Voinnet 2008; Talsky and Collins 2010).

1Corresponding authorE-mail kcollins@berkeley.eduArticle published online ahead of print. Article and publication date are

at http://www.rnajournal.org/cgi/doi/10.1261/rna.033530.112.

RNA (2012), 18:1553–1562. Published by Cold Spring Harbor Laboratory Press. Copyright � 2012 RNA Society. 1553

Cold Spring Harbor Laboratory Press on July 16, 2012 - Published by rnajournal.cshlp.orgDownloaded from

Ciliated protozoa have a large com-plexity of RNA silencing machinery rela-tive to other single-celled organisms, in-cluding numerous Piwi proteins that arethe exclusive Argonaute family membersencoded by ciliate genomes sequenced todate (Couvillion et al. 2009; Bouhoucheet al. 2011). In addition to conjugation-specific silencing pathways that inducethe transient heterochromatin formationand DNA elimination necessary for sex-ual reproduction, additional pathwaysproduce sRNAs constitutively duringculture expansion by vegetative growth(Duharcourt et al. 2009; Mochizuki 2012).The Tetrahymena Dicer, Dcr2, and thesingle genome-encoded Rdr, Rdr1, func-tion coordinately to generate multipleclasses of 23- to 24-nucleotide (nt) sRNAsin vegetatively growing cells (Lee andCollins 2006; Couvillion et al. 2009).Pseudogene transcripts, arising fromopen reading frames with downstreamgenome-encoded polyadenosine tracts,generate unphased antisense sRNAs.Other mRNA-like transcripts with po-tential to form a long 39 hairpin generatea phased register of strand-specific an-tisense sRNAs, which map in clustersupstream of the putative hairpin. Other,less abundant classes of sRNAs derivefrom repetitive regions of the genome,including telomeres, or from sites ofconvergent transcription that have po-tential to produce dsRNA by annealingrather than Rdr-mediated synthesis.

Distinct Tetrahymena RDRCs (Fig.1A) direct sRNA biogenesis from dif-ferent types of initiating transcript, withonly the RDRC-synthesized strand sRNAsstabilized by assembly with a TetrahymenaPiwi (Twi) protein. Tetrahymena RDRCsshare the Rdr1 subunit but assemble eitherRdn1 or Rdn2, one of two paralogousproteins with similar uridylyltransferaseactivity in vitro (Lee and Collins 2007;Lee et al. 2009; Talsky and Collins 2010).Rdn1 RDRCs also recruit Rdf1 and/or Rdf2, two otherparalogous proteins that lack any known protein domain(Lee and Collins 2007; Lee et al. 2009). Independent of thesedifferences in subunit composition, each RDRC copurifiesdirectly associated Dcr2 enzyme in low-salt wash conditions(Lee and Collins 2007; Lee et al. 2009). Although Dcr2, Rdr1,and Rdn1 are genetically essential for vegetative growth, eachof the other individual RDRC subunits is not (Lee and

Collins 2006, 2007; Lee et al. 2009; Malone et al. 2005;Mochizuki and Gorovsky 2005). Pseudogene-derived sRNAsare absent in cells lacking Rdn2, phased-cluster sRNAs areabsent in cells lacking Rdf2, and less abundant classes ofsRNAs are affected in more than one of the individualRDRC subunit knockout strains (Couvillion et al. 2009; Leeet al. 2009). At the cellular level, Rdf1 or Rdf2 knockout givesrise to a partially penetrant cell-division phenotype, while

FIGURE 1. Subunit sharing by RDRPs and RSPCs. (A) Schematics of the sets of proteincomplexes containing Rdr1 or Rsp1 are shown, based on previous studies and studiesdescribed here. The same symbol shapes and fill densities are used to annotate proteinidentities in the SDS-PAGE gels; RDRC-associated Dcr2 is annotated in SDS-PAGE gel panelsas an asterisk. (B) SDS-PAGE and silver staining were used to detect the subunit content ofcomplexes purified from cell extracts of the indicated tagged protein expression strain or fromwild-type cells lacking a tagged protein (mock). Complexes were purified by IgG agarosebinding and TEV protease elution. (C) A C-terminal fzz tag followed by a Neo2 drug-resistance cassette was integrated at the RSP1 locus as schematized. Locus assortment wasanalyzed by genomic DNA digestion with the indicated restriction enzyme and Southern blotdetection of differentially sized DNA fragments (the genomic region used as probe is shown).(D) SDS-PAGE and silver staining were used to detect the subunit content of complexespurified from cell extracts of the indicated tagged protein expression strain or from wild-typecells lacking a tagged protein (mock). Complexes were purified by sequential IgG agarosebinding, TEV protease elution, Flag antibody binding, and Flag peptide elution. Onepurification, Rsp1-fzz(low), was washed only in low-salt buffer.

Talsky and Collins

1554 RNA, Vol. 18, No. 8

Cold Spring Harbor Laboratory Press on July 16, 2012 - Published by rnajournal.cshlp.orgDownloaded from

loss of Rdf1 or Rdn2 blocks the progression of nuclear eventsin conjugation (Lee et al. 2009). These findings imply thateach RDRC has distinct function(s) in vivo, but in vitro eachRDRC has similar biochemical activities and Rdr1 itself issufficient for processive dsRNA synthesis (Lee et al. 2009;Talsky and Collins 2010). The Rdn subunit of an RDRC canact before Rdr1 to add mono- or oligo-uridine to a single-stranded RNA 39 end, which promotes use of the RNA 39

end in the obligate back-priming mechanism of TetrahymenaRdr1 initiation (Talsky and Collins 2010). Template tailingprior to dsRNA synthesis affects the register of cotranscrip-tional cleavage by Dcr2 (Talsky and Collins 2010).

Additional understanding of the specialization andregulation of RDRCs is needed to elucidate how particularmRNA and mRNA-like transcripts are selected for com-mitment to dsRNA synthesis in vivo. Our previous studiesshowed that although Rdn1 and Rdn2 have similar catalyticactivity, only Rdn1 is essential for culture growth (Lee et al.2009). Rdn1 but not Rdn2 interacts with an uncharacter-ized protein of 124 kDa, which we designate here as RNAsilencing protein 1 (Rsp1). Genetic, molecular, and biochem-ical assays reveal that Rsp1 is required for strand-specificantisense sRNA production by all RDRCs and suggest newsteps of sRNA biogenesis in pathways with Rdr-dependentinitial dsRNA synthesis.

RESULTS

Rsp1 interacts with RDRC subunits but not RDRCs

To further our understanding of functional distinctionsamong Tetrahymena RDRCs (Fig. 1A), we characterized the124 kDa protein detected in association with Rdn1 (Leeet al. 2009). By using strains created previously (Lee andCollins 2007; Lee et al. 2009), we confirmed that the 124-kDa protein was not evident in an IgG agarose purificationof Rdr1 tagged with tandem Protein A domains (zz-Rdr1)but was abundant in a parallel purification of Rdn1 withtandem Protein A domains and triple Flag peptide (zzf-Rdn1). Compared to a mock purification from wild-typecell extract, Rdr1 specifically enriched Rdn1, Rdn2, Rdf1,and Rdf2; Rdn1 enriched Rdr1, Rdf1, Rdf2, and a doubletof z124 kDa (Fig. 1B). Based on peptide sequences frommass spectrometry (Lee et al. 2009), we cloned the cDNAencoding the 124-kDa protein designated here as Rsp1.Despite its length, no known protein domain was detectedin Rsp1 by homology. We tagged Rsp1 at its C terminuswith a triple Flag peptide followed by tandem Protein Adomains (Rsp1-fzz), using targeted integration at the endog-enous gene locus (Fig. 1C). Based on Southern blot hybrid-ization of post-selection clonal cell lines, the transgene encod-ing Rsp1-fzz completely replaced the wild-type RSP1 locus(Fig. 1C).

We performed two-step affinity purification of RDRCs andRsp1 complexes (RSPCs) from extracts of cells expressing

tagged proteins, using wild-type cell extract as a mockpurification control. As expected, in addition to Rdr1 andDcr2, zzf-Rdn1 copurified Rsp1 and the Rdf subunits thatare not associated with zzf-Rdn2 (Fig. 1D, lanes 1,2; notethat some Dcr2 was retained through the purification ofRdn1 RDRCs even with our standard 200 mM NaCl salt-wash conditions). In comparison, Rsp1-fzz did not co-purify Rdr1 or Dcr2 but did copurify proteins that comigratewith Rdf1 and Rdf2 (Fig. 1D, lane 3). These associatedproteins were lost from Rsp1-fzz purifications in doublegene knockout background lacking Rdf1 and Rdf2 (South-ern blot data not shown), demonstrating their identity(Fig. 1D, lane 4). Even gentle low-salt wash conditionsthat stabilize RDRC-Dcr2 association did not reveal Rdr1or Dcr2 as an interaction partner of Rsp1 (Fig. 1D, lane5). We conclude that RDRCs and RSPCs share Rdn1 andthe Rdfs but are mutually exclusive for Rdr1 or Rsp1(Fig. 1A).

The purifications described above and in previous studiesdo not resolve whether the related Rdf proteins are mutuallyexclusive components of an RDRC or RSPC, as are therelated Rdn1 and Rdn2, or can be present together withina single RDRC. To address this question, we tagged the Cterminus of Rdf1, Rdf2, and Rdn2 expressed at their en-dogenous loci to generate fusion proteins with a triple Flagtag and GFP (the fg tag). Based on Southern blot hybrid-ization of post-selection clonal cell lines, the transgeneencoding each fusion protein replaced the expressed wild-type locus (Fig. 2A). Perhaps due to diffuse cytoplasmicdistribution, endogenous subunit expression levels wereinsufficient to detect subcellular concentration of the GFPsignal (data not shown). Affinity purifications of the fg-tagged proteins using Flag antibody resin recovered theexpected specificity of Rdr1, Rdn1, and Rsp1 association,without Rdf1 copurification of Rdf2 or vice versa (Fig. 2B,cf. lanes 1–4 and lane 5). Thus, like Rdn1 and Rdn2, Rdf1and Rdf2 are alternative subunits of an RDRC or RSPC(Fig. 1A).

Rdn1 activity differs in RDRC or RSPC context

Rdn1 and Rdn2 have robust uridylyltransferase activity onsingle-stranded RNA substrates when assayed as part ofa purified RDRC (Lee et al. 2009; Talsky and Collins 2010).Rdn activity as a uridylyltransferase does not depend onRdr1 activity in vitro, because it is unaffected in RDRCspurified with the catalytic-dead Rdr1(D1004A). To investi-gate whether Rdn1 retains uridylyltransferase activity inRSPC context, we used our standard Rdn assay conditionswith a 79-nt single-stranded RNA and radiolabeled a32P-UTP (Talsky and Collins 2010). Products were resolved byformamide-urea polyacrylamide gel electrophoresis (PAGE)to prevent potential RNA secondary structure formation.Compared to a mock purification from wild-type cell extract(Fig. 3A, lane 1), product radiolabeling occurred specifi-

Tetrahymena small RNA production requires Rsp1

www.rnajournal.org 1555

Cold Spring Harbor Laboratory Press on July 16, 2012 - Published by rnajournal.cshlp.orgDownloaded from

FIG

UR

E2.

Rd

f1an

dR

df2

are

mu

tual

lyex

clu

sive

par

tner

so

fR

dn

1R

DR

Cs

or

RSP

Cs.

(A)

AC

-ter

min

alfg

tag

foll

ow

edb

ya

Neo

2d

rug-

resi

stan

ceca

sset

tew

asin

tegr

ated

atth

eR

DN

2,R

DF

1,o

rR

DF

2lo

cus

assc

hem

atiz

ed.

Ass

ort

men

tat

each

locu

sw

asan

alyz

edb

yge

no

mic

DN

Ad

iges

tio

nw

ith

the

ind

icat

edre

stri

ctio

nen

zym

e(s)

and

Sou

ther

nb

lot

det

ecti

on

of

dif

fere

nti

ally

size

dD

NA

frag

men

ts(t

he

gen

om

icre

gio

ns

use

das

pro

bes

are

sho

wn

).(B

)P

rote

inco

mp

lexe

sw

ere

pu

rifi

edfr

om

cell

extr

acts

of

the

ind

icat

edta

gged

pro

tein

exp

ress

ion

stra

ino

rfr

om

wil

d-t

ype

cell

sla

ckin

ga

tagg

edp

rote

in(m

ock

)b

yF

lag

anti

bo

dy

bin

din

gan

dp

epti

de

elu

tio

n.

Talsky and Collins

1556 RNA, Vol. 18, No. 8

Cold Spring Harbor Laboratory Press on July 16, 2012 - Published by rnajournal.cshlp.orgDownloaded from

cally in purifications of Rsp1-fzz that used either the typical200 mM NaCl or gentle 50 mM NaCl wash conditions(Fig. 3A, lanes 2,3, respectively). As characterized previouslyfor RDRCs (Talsky and Collins 2010), RSPCs produced bothnear-template length and longer polyuridine products. How-ever, when normalized by the input of Rdn1 (Fig. 3, lowerpanels), RSPCs had much less specific activity than RDRCs.Compared to Rdn1 RDRCs purified using zz-Rdr1 in anRdn2 knockout background (Lee et al. 2009), the uridylyl-transferase products of RSPCs were dramatically reduced(Fig. 3B, lanes 1,2). This difference in activity was consis-tent across assays using different RNA substrates (data notshown). Mixtures of RSPCs and RDRCs copurified by taggedRdn1, Rdf1, or Rdf2 catalyzed the expected intermediatelevel of activity normalized to total input Rdn1 (Fig. 3B,lanes 3–5).

We next assayed the uridylyltransferase activity of RSPCsin comparison to a serial dilution of Rdn1 RDRCs (Fig.3C). RDRC dilution by about 1000-fold brought the levelof uridylyltransferase activity to match that of RSPCs withnormalized Rdn1 content (Fig. 3C, cf. lanes 2–5 and lane1). As a control for RSPC copurification of a nonspecificRdn1 inhibitor, the mixing of purified RSPCs and RDRCsdid not inhibit Rdn1 activity (Fig. 3C, shorter exposure oflanes 5,6). Purification of Rsp1-fzz from extracts of cellslacking Rdf1 and Rdf2 did not obviously change RSPCuridylyltransferase activity (Fig. 3D), consistent with thelack of Rdf subunit influence on the in vitro uridylyltrans-ferase activity of an RDRC (Lee et al. 2009). The lengthprofile of product synthesis by RSPCs and RDRCs was notnotably different; considering the high specific activity ofthe polyuridine addition products, shorter mono- or oligo-uridine addition products are predominant. Formally itremains possible that Rsp1 itself catalyzes uridylyltransfer-ase activity, but because catalytic-dead Rdn1 is highly toxic

for Tetrahymena growth (Lee et al. 2009), it is not possibleto assay Rsp1 activity in an RSPC with inactive Rdn1.

Cells lacking Rsp1 are viable for growth butcompromised in sexual reproduction

To address the in vivo function of Rsp1, including itspotential influence on sRNA biogenesis, we targeted theendogenous RSP1 locus for disruption with a cassette con-ferring resistance to blasticidin (see Materials and Methods).In parallel we targeted the RDN1 locus for disruption withthe same cassette. Based on Southern blot hybridization ofpost-selection clonal cell lines, the macronuclear chromo-some with disruption of RSP1 could completely replace thechromosome containing wild-type RSP1 in two strains ofcomplementary mating type (CU522 clonal strain 4 andSB210 clonal strain 4 were used for subsequent studies)(Fig. 4A). Thus, as is true for its associated Rdf1 and Rdf2subunits, Rsp1 is not essential for vegetative growth. Incontrast, consistent with previous results (Lee et al. 2009),the wild-type RDN1 locus could not be fully replaced (datanot shown).

Conjugating cell pairs that lack parental Rdn2 or Rdf1fail to progress through the nuclear differentiation eventsthat produce new macronuclei (Lee et al. 2009). We thereforeaddressed whether cells lacking parental Rsp1 also havea conjugation defect. Mated wild-type SB210 and CU522cells formed stable pairs and by 9 h post-mixing generatedtwo new developing macronuclei and two new nuclei withmicronuclear content positioned anterior to the stronglyDAPI-staining, posterior-localized parental macronucleus(Fig. 4B, top panels; the anterior ends of the paired cellsare indicated with arrowheads). In contrast, although stablypaired and normal compared with wild-type cells for z4 hpost-mixing, mated SB210 and CU522 cells with macronu-

FIGURE 3. Rdn1 activity is influenced by association with Rdr1 or Rsp1. (A–D) Protein complexes were purified from extracts of wild-typecells or the indicated tagged protein expression strain by binding to IgG agarose and elution with TEV protease. One purification, Rsp1-fzz(low), was washed only in low-salt buffer. Rdn1 protein content was normalized by silver staining after SDS-PAGE (bottom panels), prior toactivity assay by radiolabeled UTP addition to the single-stranded RNA1. Polyuridine addition products at the top of the gel are resolved fromthe less processively elongated products migrating at approximately input template length. The same purifications used in B were also used fordilution or mixture in C.

Tetrahymena small RNA production requires Rsp1

www.rnajournal.org 1557

Cold Spring Harbor Laboratory Press on July 16, 2012 - Published by rnajournal.cshlp.orgDownloaded from

clear RSP1 knockout accumulated with a mid-cell parentalmacronucleus and several smaller, presumably micronuclear-content zygotic nuclei (Fig. 4B, bottom panel). This stage ofconjugation arrest is similar to that observed upon matingRDF1 or RDN2 macronuclear gene knockouts (Lee et al.2009). However, at low frequency, occasional RSP1 macro-nuclear knockout cell pairs appeared to progress past this

early conjugation arrest. We note that loss of Rsp1 couldimpose a conjugation defect indirectly through an impact onthe stability and/or function of Rdf1, Rdn1, and/or Rdn2.

Accumulation of strand-specific antisense sRNAsrequires both Rdr1 and Rsp1

The high abundance of 23- to 24-nt sRNAs in growingTetrahymena allows their direct visualization by filtration-based size enrichment, denaturing gel electrophoresis, andSYBR Gold staining (Lee and Collins 2006). To investigatewhether the loss of Rsp1 function influences the accumu-lation of these Dcr2-generated sRNAs, we analyzed size-selected RNA from strains lacking Rsp1 or various RDRCsubunit(s). In parallel we analyzed size-selected RNA fromthe corresponding wild-type strain and a strain that induciblyoverexpressed catalytic-dead Rdr1(D1004A) in competitionwith endogenous Rdr1 (Lee and Collins 2007). Surpris-ingly, cells lacking Rsp1 also lacked abundant 23- to 24-ntsRNAs (Fig. 5, cf. lanes 1 and 2). This loss was even morestriking than the depletion of 23- to 24-nt sRNAs in anyindividual or combination knockout of the nonessentialRDRC subunits (Southern blots [data not shown] and sRNAphenotypes [see below] confirm the knockout genotypes)(Fig. 5, lanes 4–8). Overexpression of catalytic-dead Rdr1was also effective in depleting 23- to 24-nt sRNAs, perhapswith an increase in the general RNA fragment background(mature tRNA provides a control for RNA integrity andrelative loading) (Fig. 5, lane 3).

We next used blot hybridization with individual sRNA-complementary oligonucleotides to detect representativemembers of each abundant Tetrahymena class of sRNA(Fig. 5, lower panels), based on previous deep sequencingand blot hybridization data (Couvillion et al. 2009). Asexpected from the depletion of bulk 23- to 24-nt sRNAs,cells lacking Rsp1 or expressing catalytic-dead Rdr1 weremissing the sRNA classes associated with Twi2. These sRNAclasses include the pseudogene loci that generate sRNAdependent on the Rdn2 RDRC (e.g., the IIIB locus), phasedcluster loci that generate sRNA dependent on the Rdf2RDRC (e.g., the PH3 locus), and loci that generate sRNAdependent on more than one type of RDRC (e.g., the IBlocus and high-copy repeat RPT1). Of note, a primary tran-script capable of extensive secondary structure formation stillgenerated processed RNA even in Rsp1 knockout cells orcells expressing catalytic-dead Rdr1 (Fig. 5, STR1), althoughwith a broadened sRNA size distribution. This population ofsRNA was most enriched in association with nuclear Twi8rather than cytoplasmic Twi2 (Couvillion et al. 2009) andcould be generated directly by Dcr2 cleavage of an annealeddsRNA hairpin.

Coincident with the loss of phased-cluster sRNAs inRdf2 knockout strains, longer transcripts that could beRdf2 RDRC precursors become detectable (Couvillion et al.2009). For the PH3 locus, this longer transcript migrates as

FIGURE 4. Loss of Rsp1 is tolerated in vegetative growth butimpedes conjugation. (A) A drug-resistance cassette was integratedat the RSP1 locus as schematized in strains with different matingtypes. Assortment was analyzed by genomic DNA digestion with theindicated restriction enzyme and then Southern blot detection ofdifferentially sized DNA fragments (the genomic region used as probeis shown). Asterisks indicate the clonal cell lines used for subsequentstudies. (B) Cells were fixed at the 9-h time point of conjugation andstained with DAPI. Each conjugating cell pair (indicated by pairedarrowheads) represents the predominant state of nuclear differentia-tion observed in wild-type cells (top panels) or Rsp1 knockout cells(bottom panel). In wild-type cell pairs, the two developing zygoticmacronuclei stain lightly, the parental macronucleus stains strongly,and two smaller micronuclear-content genomes are also evident. Cellpairs lacking parental Rsp1 retain the parental macronucleus. Anunpaired cell with a single macronucleus and single micronucleus isindicated by an asterisk. All panels are shown at the same scale; scalebar, 10 mm.

Talsky and Collins

1558 RNA, Vol. 18, No. 8

Cold Spring Harbor Laboratory Press on July 16, 2012 - Published by rnajournal.cshlp.orgDownloaded from

>500 nt when detected by mRNA Northern blot (Fig. 6A).To investigate the fate of a putative RDRC substrate inRsp1 knockout cells, we first defined transcripts from thePH3 locus at a molecular level using RT-PCR. Transcriptscontained the two regions that give rise to clustered sRNAs,joined by mRNA splicing and followed by a previouslypredicted 163-nt stem–loop and at least 300 nt of sequencebeyond the structured region (Fig. 6B). Transcript sequenceswere readily amplified by PCR after reverse transcriptionwith dT18 primer, which suggests that at least some of thetranscript population becomes polyadenylated. BecauseRDRC initiation requires a free single-stranded RNA 39 end(Talsky and Collins 2010), the direct precursor for RDRC

dsRNA synthesis may be generated by endonucleolyticcleavage of the polyadenylated transcript (Fig. 6B).

To investigate potential RSPC involvement in the gen-eration of an RDRC precursor, we probed for PH3 transcriptin cells lacking Rsp1. In contrast to cells lacking Rdf2, cellslacking Rsp1 did not accumulate an mRNA-sized PH3transcript in either of two strain backgrounds (Fig. 6A,cf. lanes 3,4 and 7,8). Interestingly, the mRNA-like PH3 tran-script also did not accumulate in cells expressing catalytic-dead Rdr1 (Fig. 6A, lane 8). Because cells lacking Rsp1 orexpressing catalytic-dead Rdr1 also did not accumulate PH3

FIGURE 5. Rdr1-dependent sRNA accumulation requires Rsp1. Size-selected RNA from the indicated strains was analyzed by denaturingPAGE and SYBR Gold stain (top panel). Northern blot hybridizationswere performed with probes antisense to sRNAs IIIB, IB, PH3, RPT1,and STR1 (bottom panels), which were previously sequenced inenriched association with Twi2 or Twi8 as indicated. The same blotwas probed for all sRNAs except for STR1, for which a separate blotcontaining the same RNAs was probed.

FIGURE 6. PH3 precursor transcripts do not accumulate in cellslacking Rsp1. (A) Detection of a PH3 locus transcript by Northernblot using hexamer-primed probe. The ribosomal RNA (rRNA)region of total RNA stained by ethidium bromide is shown asa loading control; note that endogenous cleavage of Tetrahymena26S rRNA creates a doublet that overlaps in migration with 17S rRNA.(B) PH3 mRNA-like transcript contains two exons (solid black lines)separated by an intron (dashed line). At least some fraction of PH3transcript is processed as mRNA, based on RT-PCR detection ofintron removal and polyadenylation. Further processing may occurbefore recognition of the transcript as an RDRC substrate, forexample, a speculated endonucleolytic cleavage at the base of thestem–loop to generate a nonpolyadenylated 39 end.

Tetrahymena small RNA production requires Rsp1

www.rnajournal.org 1559

Cold Spring Harbor Laboratory Press on July 16, 2012 - Published by rnajournal.cshlp.orgDownloaded from

sRNA, the mRNA-sized transcript may have escaped com-mitment to sRNA biogenesis in a manner that allowed analternative pathway of mRNA decay. Several models forRsp1 function can be envisioned, including potential Rsp1activities prior to Rdr1 in RNA silencing pathways (Fig. 7;see Discussion).

DISCUSSION

Together the results above suggest that Rsp1 plays a role inRDRC-mediated sRNA biogenesis. Rsp1 assembles stablecomplexes with Rdn1 and an Rdf that are mutually exclu-sive for binding of Rdr1 and Dcr2. Consistent with thephysical segregation of RSPCs and RDRCs, RSPCs lackassociated RDRC activity (data not shown). However, cellslacking Rsp1 or expressing catalytic-dead Rdr1(D1004A)were similarly depleted for all known classes of Twi2-boundsRNAs. Curiously, Rdn1 uridylyltransferase activity wasmarkedly reduced when bound to Rsp1 compared with Rdr1.This inhibition could be important for limiting Rdn1activity in vivo.

Models for Rsp1 function include potential roles beforeor after dsRNA synthesis. RSPCs could select RNAs for

commitment to an RDRC pathway (Fig. 7, model a). RNAselection or processing by an RSPC-associated activity, forexample, an mRNA endonuclease, could provide the tem-plate specificity that an RDRC itself may lack. The subunitsharing between RSPCs and RDRCs may directly or in-directly facilitate RNA handoff. Alternately, RSPCs couldmaintain a cellular reservoir of Rdn1/Rdf complexes that issequestered from Rdr1 or that promotes stable Rdn1/Rdffolding prior to RDRC assembly (Fig. 7, model b). The mod-els above do not necessarily predict the observed influenceof Rsp1 loss of function on Rdn2 RDRC sRNAs. Insteadof reflecting a loss of function, the disruption of Rdn2-dependent pseudogene loci sRNA biogenesis could reflecta dominant-negative impact of altered RNA silencing. In anentirely different type of model, Rsp1 could act subsequentto RDRC-mediated dsRNA synthesis and cotranscriptionalcleavage by Dcr2 (Fig. 7, model c). For example, RSPCscould modify sRNAs or mediate sRNA loading to Twi2,thereby promoting sRNA stabilization as RNP. However,if RSPC function is downstream from Dcr2, it is notobvious why RSPCs and RDRCs would share Rdn1 andRdf subunits.

To our surprise, we found that unlike Rdr1 and Rdn1(Lee and Collins 2007; Lee et al. 2009), Rsp1 is not essentialfor vegetative growth. The corresponding viability of cellslacking any known Twi2-bound sRNA suggests that pro-duction of these sRNAs may be physiologically mostsignificant as the degradation end-point for nonfunctionaltranscripts rather than as a mechanism of post-transcriptionalgene silencing. This would be consistent with our observa-tions that some of the 23- to 24-nt sRNA biogenesis machin-ery have a genetically essential role in conjugation ratherthan vegetative growth (Lee et al. 2009). The need for Rdn1,Rdr1, and Dcr2 function in vegetative growth could reflecta requirement to prevent the accumulation of toxic in-termediates of sRNA biogenesis. Aberrantly accumulatingmRNA or mRNA-like transcripts could overwhelm thecellular machinery for RNA surveillance and degradation,or accumulating dsRNAs could constitutively activate viraldefense (Howard-Till and Yao 2006).

The apparently RDRC-independent biogenesis of at leastone class of Twi8-bound sRNAs suggests that these sRNAscould be generated by direct Dcr2 cleavage, uncoupled fromdsRNA synthesis by Rdr1. Curiously, in vitro Dcr2 cleavageof annealed dsRNA can generate an increased heterogeneityof sRNA lengths compared to cleavage that is cotranscrip-tional with dsRNA synthesis (Lee and Collins 2007). Corre-spondingly, cells lacking Rsp1 or expressing catalytic-deadRdr1(D1004A) have an increased heterogeneity of STR1sRNAs. The more uniform size distribution of STR1 sRNAsin wild-type cells suggests that some transcripts from locithat generate Twi8-bound sRNAs are normally directed toRDRC-mediated dsRNA synthesis and cotranscriptional Dcr2cleavage. Twi8 is not essential in at least some strain back-grounds (Couvillion et al. 2009), but it remains a possibility

FIGURE 7. Models for Rsp1 function in RDRC-dependent sRNAbiogenesis. Hypothetical Rsp1 functions (dashed arrows) are sug-gested in the context of known pathway steps (solid black arrows). Fordescription of the three models indicated as a, b, and c, see theDiscussion.

Talsky and Collins

1560 RNA, Vol. 18, No. 8

Cold Spring Harbor Laboratory Press on July 16, 2012 - Published by rnajournal.cshlp.orgDownloaded from

that some loci can produce essential 23- to 24-nt sRNAs bymore than one biogenesis pathway.

MATERIALS AND METHODS

Cell lines and culture conditions

Integration constructs for protein tagging fused the C-terminalamino acid of the endogenous open reading frame to the epitopetag(s), followed by the RPL29 39 untranslated region (UTR) anda Neo2 cassette with the BTU2 39 UTR. A protease cleavage siteseparated the tandem Protein A module from the remainder of thepolypeptide to allow elution from IgG resin (see below). Integrationconstructs for gene knockout replaced the protein open readingframe with the Bsr2 cassette (MTT1 promoter, open reading frameencoding resistance to blasticidin, BTU2 39 UTR). Cells were selectedfor integration and assortment of Neo2 using paromomycin or forintegration of Bsr2 using blasticidin with 0.7 mg/mL CdCl2 followedby assortment in blasticidin without CdCl2 to reduce cassetteexpression.

Cultures were grown shaking at 30°C in NEFF (0.5% proteosepeptone, 0.5% yeast extract, 1% glucose, 10 mM FeCl3) and harvestedfor extract production at a density of 2 3 105 to 5 3 105 cells/mL.For zz-Rdr1(D1004A) expression (Lee and Collins 2007), cells wereinoculated at 1 3 104 to 2 3 104 cells/mL and grown for 16–20 h inthe presence of 1 mg/mL CdCl2. For conjugation, cells were starved in10 mM Tris-HCl (pH 7.5), mixed, and incubated at 30°C withoutshaking. At least 200 cell pairs of each genotype were counted toscore the stage of nuclear differentiation. Cells were fixed in 2%paraformaldehyde for 1 h, washed twice in modified PBS (130 mMNaCl, 2 mM KCl, 8 mM Na2HPO4, 2 mM KH2PO4, 10 mM ethyleneglycol tetra-acetic acid at pH 7.2) with 0.1% bovine serum albumin(Sigma), incubated in 0.1 mM DAPI for 10 min, washed once morein modified PBS, and mounted in modified PBS with 70% glycerol.

Protein purification and activity assays

Proteins were purified by binding to IgG agarose and/or M2 Flagantibody resin using previously developed cell lysis and purifica-tion conditions (Lee and Collins 2007). Resin-bound complexeswere washed in 200 mM NaCl, except for the low-salt washes of50 mM NaCl as indicated. IgG agarose elution was performed byaddition of Tobacco Etch Virus (TEV) protease; Flag antibody elutionwas performed using triple Flag peptide (Sigma). Uridylyltransferaseactivity was assayed using a 79-nt single-stranded RNA1 templateand radiolabeled UTP diluted to a final concentration of 20 mM(Talsky and Collins 2010). To prevent RNA secondary structureformation during gel electrophoresis, radiolabeled products wereanalyzed by highly denaturing formamide-PAGE (9%–12% of 19:1acrylamide:bis-acrylamide, 7 M urea, 45% formamide, 13 TBE).

RNA purification and detection

RNA isolation, sRNA enrichment, and blot hybridization methodswere as previously described (Couvillion et al. 2009; Lee and Collins2006). Total RNA was extracted using TRIzol reagent (Invitrogen)from cells washed into 10 mM Tris-HCl (pH 7.5). Enrichment forsRNA was done using YM50 Microcon columns (Amicon). Blothybridization for sRNAs after acrylamide gel electrophoresis used

end-labeled DNA oligonucleotides, while hybridization for mRNANorthern blots and genomic DNA Southern blots used hexamer-labeled double-stranded DNA probes. Oligonucleotide probe se-quences were complementary to previously sequenced sRNAs: IIIB(AGCAAAGACGATTAACAATATTCA), IB (ATTTACTAGATGTATTTCCCTTA), PH3 (AACTATTTAATAATATATTTTCA), RPT1(AATATCACAATCCAAAACAAATA), and STR1 (AAGCTTCTCTTATCTTCTAATCA). The mRNA-sized PH3 transcript was an-alyzed by RT-PCR from total RNA of CU522 wild-type and Rdf2knockout cells after treatment with RQ1 RNase-free DNase(Promega).

DATA DEPOSITION

The GenBank accession number for RSP1 is JQ797583.

ACKNOWLEDGMENTS

We thank Mary Couvillion for reagents, discussion, and com-ments on the manuscript.

Received March 28, 2012; accepted May 22, 2012.

REFERENCES

Bouhouche K, Gout JF, Kapusta A, Betermier M, Meyer E. 2011.Functional specialization of Piwi proteins in Paramecium tetraur-elia from post-transcriptional gene silencing to genome remodel-ling. Nucleic Acids Res 39: 4249–4264.

Carthew RW, Sontheimer EJ. 2009. Origins and mechanisms ofmiRNAs and siRNAs. Cell 136: 642–655.

Couvillion MT, Lee SR, Hogstad B, Malone CD, Tonkin LA,Sachidanandam R, Hannon GJ, Collins K. 2009. Sequence, bio-genesis, and function of diverse small RNA classes bound to thePiwi-family proteins of Tetrahymena thermophila. Genes Dev 23:2016–2032.

Duharcourt S, Lepere G, Meyer E. 2009. Developmental genomerearrangements in ciliates: a natural genomic subtraction mediatedby non-coding transcripts. Trends Genet 25: 344–350.

Grewal SI. 2010. RNAi-dependent formation of heterochromatin andits diverse functions. Curr Opin Genet Dev 20: 134–141.

Howard-Till RA, Yao MC. 2006. Induction of gene silencing byhairpin RNA expression in Tetrahymena thermophila revealsa second small RNA pathway. Mol Cell Biol 26: 8731–8742.

Ketting RF. 2011. The many faces of RNAi. Dev Cell 20: 148–161.Lee SR, Collins K. 2006. Two classes of endogenous small RNAs in

Tetrahymena thermophila. Genes Dev 20: 28–33.Lee SR, Collins K. 2007. Physical and functional coupling of RNA-

dependent RNA polymerase and Dicer in the biogenesis ofendogenous siRNAs. Nat Struct Mol Biol 14: 604–610.

Lee SR, Talsky KB, Collins K. 2009. A single RNA-dependent RNApolymerase assembles with mutually exclusive nucleotidyl trans-ferase subunits to direct different pathways of small RNA bio-genesis. RNA 15: 1363–1374.

Malone CD, Anderson AM, Motl JA, Rexer CH, Chalker DL. 2005.Germ line transcripts are processed by a Dicer-like protein thatis essential for developmentally programmed genome re-arrangements of Tetrahymena thermophila. Mol Cell Biol 25:9151–9164.

Moazed D. 2011. Mechanisms for the inheritance of chromatin states.Cell 146: 510–518.

Mochizuki K. 2012. Developmentally programmed, RNA-directedgenome rearrangement in Tetrahymena. Dev Growth Differ 54:108–119.

Tetrahymena small RNA production requires Rsp1

www.rnajournal.org 1561

Cold Spring Harbor Laboratory Press on July 16, 2012 - Published by rnajournal.cshlp.orgDownloaded from

Mochizuki K, Gorovsky MA. 2005. A Dicer-like protein inTetrahymena has distinct functions in genome rearrangement,chromosome segregation, and meiotic prophase. Genes Dev 19:77–89.

Pak J, Fire A. 2007. Distinct populations of primary andsecondary effectors during RNAi in C. elegans. Science 315:241–244.

Sijen T, Steiner FA, Thijssen KL, Plasterk RH. 2007. Secondary siRNAsresult from unprimed RNA synthesis and form a distinct class.Science 315: 244–247.

Talsky KB, Collins K. 2010. Initiation by a eukaryotic RNA-dependentRNA polymerase requires looping of the template end and is

influenced by the template-tailing activity of an associated uridyl-transferase. J Biol Chem 285: 27614–27623.

Thivierge C, Makil N, Flamand M, Vasale JJ, Mello CC, WohlschlegelJ, Conte D Jr, Duchaine TF. 2012. Tudor domain ERI-5 tethers anRNA-dependent RNA polymerase to DCR-1 to potentiate endo-RNAi. Nat Struct Mol Biol 19: 90–97.

Vazquez F, Vaucheret H, Rajagopalan R, Lepers C, Gasciolli V,Mallory AC, Hilbert JL, Bartel DP, Crete P. 2004. Endogenoustrans-acting siRNAs regulate the accumulation of ArabidopsismRNAs. Mol Cell 16: 69–79.

Voinnet O. 2008. Use, tolerance and avoidance of amplified RNAsilencing by plants. Trends Plant Sci 13: 317–328.

Talsky and Collins

1562 RNA, Vol. 18, No. 8

Cold Spring Harbor Laboratory Press on July 16, 2012 - Published by rnajournal.cshlp.orgDownloaded from